The New Hampshire Department of Transportation (DOT) appeals the decision of the Superior Court (McGuire, J.) vacating the New Hampshire Department of Environmental Services’ (DES) approval of DOT’s application for a wetlands permit, and the New Hampshire Wetlands Council’s (Wetlands Council) affirmance of the same. We reverse and remand.
The trial court provided the following background facts in its order. This case involves the planned expansion by DOT of Routes 9,10,12 and 101 in Swanzey and Keene. The project includes the proposed replacement of the “T”. intersection at which these routes intersect with a trumpet interchange to handle greater traffic volume.
Constructing the trumpet interchange would require filling approximately 5.45 acres of wetlands. Accordingly, DOT applied to DES for a wetlands permit on June 7, 1999. See RSA 482-A:3 (2001) (amended 2001, 2002). DES granted the permit, subject to certain conditions, on March 22, 2000. The plaintiff, Conservation Law Foundation (CLF), requested reconsideration of DES’ decision under RSA 482-A:10, II (2001). DES denied the request and CLF appealed to the Wetlands Council. See *3RSA 482-A.10, IV (2001). The Wetlands Council affirmed DES’ decision and denied CLF’s motion for reconsideration. CLF then appealed to the superior court. See RSA 482-A:10, VIII (2001).
The trial court vacated the Wetlands Council’s and DES’ orders. The court stated that, under DES’ Wetlands Rules, an applicant for a wetlands permit “must demonstrate, among other things, the need for the proposed project’s impact on wetlands, and that the proposed project is the one with the least adverse impact to wetlands.” See N.H. Admin. Rules, Wt 302.03, 302.04(a) (1), (2). One alternative to the trumpet interchange was a roundabout, which the court described as follows: “A roundabout is designed as a one-way circulatory roadway around a central island usually with flared approaches to allow multiple vehicle entry. Modern roundabouts have two fundamental design features including yield at entry points and deflection of the entering vehicle path.” (Quotation omitted.) The court held, however, that even assuming the accuracy of DOT’s traffic predictions for 2015, “its application did not adequately address whether a roundabout was a practicable alternative to the proposed trumpet interchange.”
The court also held that DOT did not adequately address the planned construction’s impact on wildlife because it failed to “discover and document... vernal and seasonal pools in the area of the proposed trumpet interchange.” Vernal and seasonal pools are temporary bodies of water that provide many species a predator-free environment in which to reproduce. The court further held that the imposition of conditions in the permit did not cure the unlawfulness and unreasonableness of the permit’s issuance.
DOT appeals, arguing that the trial court erred in: 1) supplementing the record on appeal; 2) finding the discovery of vernal pools to be significant new information requiring vacating of the wetlands permit; 3) finding DES’ reliance on the conditions imposed in the permit to be unreasonable; and 4) interpreting administrative rules to outweigh statutory authority specifically delegated to a State agency.
We first set forth the applicable standards of review. The trial court’s review of decisions of the Wetlands Council is governed by RSA 482-A:10, XI (2001), which provides:
On appeal to the superior court, the burden of proof shall be upon the party seeking to set aside the decision of the council to show that the decision is unlawful or unreasonable. The council’s decision shall not be set aside or vacated, except for errors of law, unless the court is persuaded, by a preponderance of the evidence before it, that said decision is unjust or unreasonable.
*4We will not disturb the trial court’s decision unless it is unsupported by the evidence or legally erroneous. Cf. Mt. Valley Mall Assocs. v. Municipality of Conway, 144 N.H. 642, 647 (2000).
DOT first argues that the trial court erred in supplementing the record with the opinion testimony of Dr. Rick Van de Poll, who visited the site on behalf of CLF and discovered the vernal and seasonal pools at issue. As Dr. Van de Poll did not discover the vernal and seasonal pools until after the wetlands permit issued, and CLF’s previous motion to supplement the record before the Wetlands Council with his affidavit was denied, Dr. Van de Poll’s opinion had not been before either DES or the Wetlands Council.
Under RSA 482-A:10, XIII (2001), the trial court hearing an appeal from the Wetlands Council is directed to order the Council to file with the court a certified copy of the record, which “shall consist of the council’s decision, [DES’] record of decision as submitted to the council and the record of the hearing before the council.” Nevertheless, RSA 482-A:10, XIV (2001) provides that “[t]he court may receive and consider such additional evidence as would be permissible under RSA 677:10.”
We have held:
The purpose of the statutory provisions for the receipt of additional evidence is not to afford a trial de novo, but rather to assist the court in evaluating the action of the board. If the record of the evidence before the board is either incomplete or nonexistent, a party may present further evidence to the trial court[, which may be considered] even though it was not before the board.
Lake Sunapee Protective Assoc. v. N.H. Wetlands Bd., 133 N.H. 98, 106-07 (1990) (quotations, citations, brackets and ellipses omitted). The decision to admit additional evidence is within the trial court’s discretion and we will not reverse the court’s decision absent an unsustainable exercise of discretion. See Peter Christian’s v. Town of Hanover, 132 N.H. 677, 683-84 (1990); State v. Lambert, 147 N.H. 295, 296 (2001).
DOT contends that the trial court erred in taking additional evidence because the record was neither nonexistent nor incomplete. Rather, DOT argues that the record below was “voluminous” and contained “the culmination of substantial environmental research on behalf of [DOT] and input from” numerous agencies and others. CLF, on the other hand, argues that the record before DES and the Wetlands Council was necessarily incomplete because it contained no reference to the subsequently-discovered vernal and seasonal pools.
*5We have recognized that “the trial judge is in the best position to determine the sufficiency of the record.” Peter Christian’s, 132 N.H. at 683-84. Thus, although we ultimately conclude that the evidence of vernal and seasonal pools did not have the significance the trial court accorded it, we cannot say that the court’s decision to admit the evidence was an unsustainable exercise of discretion. See id.; Lambert, 147 N.H. at 296.
Having admitted the evidence of vernal and seasonal pools, the trial court concluded that DOT’s failure to find and document them constituted a failure to adequately consider the project’s impact on wildlife in accordance with New Hampshire Administrative Rule, Wt 302.04(a)(7). Rule 302.04(a) provides, in part:
(a) For all major and minor projects the applicant shall demonstrate by plan and example that the following factors have been considered in their design in assessing the impact of the proposed project to areas and environments under the department’s jurisdiction:
(7) The impact on plants, fish, and wildlife including:
a. Rare, special concern species;
b. State and federally listed threatened and endangered species;
c. Species at the extremities of their ranges;
d. Migratory fish and wildlife; and
e. Exemplary natural communities identified by the New Hampshire Natural Heritage Inventory (NHI) — Department of Resources and Economic Development.
The trial court held that the Wetlands Council erred in interpreting that provision “as requiring applicants to assess impacts to rare, threatened and endangered and migratory fish and wildlife and NHI communities; no other wildlife concerns are included in the rules.” (Quotation omitted.) The court concluded that the Wetlands Council’s interpretation was too narrow and ignored the use of the term “including,” which indicated that the list of enumerated factors was not exhaustive. It then stated: “While vernal and seasonal pools are not specifically mentioned in the rules, their existence is critical to understanding the exact nature and character of the proposed project’s impact on wildlife habitat, reproduction, and food sources.”
While we agree with the trial court that the term “including” indicates that the factors listed are not exhaustive, we conclude that the trial court’s interpretation of the rule to require documentation of vernal and seasonal pools in all cases is too broad. In Roberts v. General Motors Corp, 138 N.H. *6532, 538 (1994), we noted that the phrase “including but not limited to,” when used in a statute preceding a list of specified items, “limits the applicability of [the statute] to those types of [items] therein particularized.” Id. at 538 (quotation omitted).
We hold that the term “including” similarly limits the items intended to be covered by the rule to those of the same type as the items specifically listed. Thus, Rule 302.04(a)(7) requires consideration of the proposed project’s impact on certain plants, fish and wildlife that are of particular concern for conservation, wildlife management or regulatory purposes. While vernal and seasonal pools may provide habitat and other benefits to certain species, without evidence that the project’s impact on the pools would adversely affect specific species contemplated by Rule 302.04(a)(7), we cannot say that the rule requires documentation of vernal or seasonal pools.
The dissent argues that Rule 302.04(a)(7) should be interpreted in light of the purpose expressed in RSA 482-A-.1, which is, among other things, to protect “sources of nutrients for finfish, Crustacea, shellfish and wildlife of significant value” and “habitats and reproduction areas for plants, fish and wildlife of importance.” RSA 482-A:l (2001). We see no substantive difference between the stated limitations “of significant value” and “of importance,” and our above interpretation “of particular concern.”
Craig Wood, manager of environmental services for the Louis Berger Group, which prepared a study for DOT of the vernal pools at issue, testified that their investigation uncovered no rare, special concern, threatened or endangered species, species at the extremities of their ranges, or migratory fish or wildlife. He also testified that none of the species identified by Dr. Van de Poll as occurring in these pools fit any of the categories listed in Rule 302.04(a)(7). He also stated that vernal pools themselves are not rare; nor did these pools provide any rare functions.
Both Dr. Van de Poll and Mr. Wood testified that the vernal and seasonal pools were within the wetlands delineated in DOT’s application and that their existence did not change the principal function of the wetlands as listed by DOT, which was wildlife habitat. Mr. Wood also testified that the discovery of vernal pools at the site would have no effect on the analysis of the least environmentally damaging practicable alternative. Based on the foregoing, we conclude that the trial court’s interpretation of Rule 302.04(a)(7) was legally erroneous and its determination that the existence of vernal pools required vacating the wetlands permit was unsupported by the evidence. Cf. Mt. Valley Mall, 144 N.H. at 647. Because we conclude that the trial court erred in vacating *7the decisions of DES and the Wetlands Council on this basis, we need not address whether the court erred in finding that DES’ imposition of conditions in the permit, which, among other things, required an updated assessment of the project’s impact on wetlands, did not cure the unreasonableness and unlawfulness of the permit’s issuance.
We next address the trial court’s ruling that DOT failed to demonstrate: 1) the need for the project’s proposed impact on wetlands; and 2) that the project is the alternative with the least adverse impact to wetlands. See N.H. Admin. Rules, Wt 302.03, 302.04(a) (1), (2). Specifically the court held that DOT “did not adequately address whether a roundabout was a practicable alternative to the proposed trumpet interchange.”
The court stated that DOT rejected a multi-lane roundabout as an alternative because it did not meet capacity needs for the projected 2015 traffic flow. The court concluded, however, that although DOT conducted three engineering assessments of roundabouts, it “failed to consider geometric design elements, which greatly increase the vehicles per hour which a roundabout can accommodate.”
The court relied on studies contained in the record that emphasized the importance of geometry in roundabout design. The court stated:
The process of designing roundabouts, more so than other forms of intersections, requires a considerable amount of iteration between geometric layout, operational analysis, and safety evaluation----[M]inor adjustments in geometry can result in significant changes in the safety and/or operational performance. Thus, the designer often needs to revise and refine the initial layout attempt to enhance its capacity and safety. It is rare to produce an optimal geometric design on the first attempt.
(Quotation omitted.) The court also noted that a report from the United Kingdom stated that roundabouts could be designed to accommodate 6,000 to 8,000 vehicles per hour, a range that included the 6,455 vehicles per hour predicted by DOT for peak afternoon traffic flow in 2015.
We conclude that the trial court’s finding that DOT failed adequately to address whether a roundabout was a practicable alternative because it did not consider geometric design elements is unsupported by the evidence. Gilbert Rogers, DOT’S assistant commissioner and chief engineer, admitted that DOT did not fine-tune a roundabout design, but testified that DOT did consider roundabouts “enough to know whether [the] concept [was] workable or not.” Specifically, he testified that in his professional opinion, a three-lane roundabout would be required at the location of the proposed trumpet interchange in order to accommodate the *8project’s demands. He also testified that the document DOT uses for roundabout design only recommends two-lane roundabouts. In addition, in a letter to the chairman of the Keene Advisory Task Force, Carol Murray, assistant commissioner of DOT, and Kathleen Laffey, Division Administrator of the Federal Highway Administration, stated that “[sjinee roundabouts are a newer design feature, the public’s likely ability to adapt to using them is an important consideration.” They then noted that “three lane roundabouts are too complicated for the average American driver to negotiate.”
The evidence before the trial court indicated that alteration of the geometric design would not solve the three-lane problem. Thus, while Mr. Rogers agreed that slip lanes or bypass lanes are fine-tuning adjustments that are available but “were probably not developed” in DOT’s alternatives analysis, he explained that those lanes merely allow right-turning traffic to turn without having to enter the roundabout, and do not affect “how many vehicles can circulate and parallel around the circle at one time,” or, in other words, how many lanes the roundabout must have. He concluded that “you can do all the fine tuning adjustments you wanted to, but the problem is a three-lane roundabout is not the appropriate answer.”
Mr. Roger’s testimony on this point is corroborated by a graph in plaintiffs exhibit K that charts vehicle entry flow against circulating flow. Certain areas of the graph indicate the use of a single-lane, two-lane, or three-lane roundabout, with overlap areas shown as “shaded bands [which] indicate conditions in which either treatment may be suitable depending on the geometry and acceptable operating conditions.” DOT’s projected traffic volumes chart from the upper bound of the shaded area between two- and three-lane roundabouts (for what appears to be east-bound traffic) to well into the three-lane area (for what appears to be west- and south-bound traffic). Thus, DOT’s analysis showed that no amount of geometric iteration would take a roundabout at this location out of the unacceptable three-lane range. Based on the foregoing, we cannot agree with the trial court that “DOT was obligated to design roundabouts manipulating the various geometric variables to determine whether a roundabout is a practicable alternative to the proposed trumpet interchange.”
Finally, CLF argues that even if DOT prevailed on any of its issues on appeal, the trial court made numerous other findings of fact and rulings of law, which DOT has not challenged on appeal, that invalidate its permit. We decline to address this argument on appeal. Although the trial court made rulings on proposed findings of fact and rulings of law that may appear inconsistent with DES’ and the Wetland’s Council’s decisions, it did *9not base its decision upon them. Rather, it expressly stated that because it found the Wetlands Council’s decision unlawful and unreasonable for the reasons stated, it “need not address the CLF’s remaining arguments.” We will not speculate as to the effect these findings and rulings might have had upon the trial court’s decision had it chosen to address them in its order. Thus, we decline to consider CLF’s argument.
For the foregoing reasons, we reverse the trial court’s decision and remand to the superior court.
Reversed and remanded,
Broderick and Dalianis, JJ., concurred; Duggan, J., dissented.